Background: Neuroendocrine neoplasms (NENs) encompass a rare group of tumors arising in nearly any organ site. Patients with advanced-stage NENs have limited treatment options, and molecular profiling is not routinely performed in NENs. However, given tissue-agnostic FDA approvals for immunotherapy in biomarker-selected solid tumors, we sought to assess the frequency of TMB-H and MSI-H in NENs by organ site and in comparison to non-NEN counterparts.

Methods: We retrospectively analyzed de-identified records from the Tempus database of NENs sequenced with the Tempus xT assay (DNA-seq of 595-648 genes at 500x coverage). Patients were included if their primary cancer arose in the lungs, GI tract (esophagus, stomach, small bowel, colon, rectum), pancreas, or prostate. Histologic diagnoses were grouped as NENs (including high grade or low/intermediate grade), and non-NENs (e.g. adenocarcinoma, squamous cell carcinoma) based on data abstracted from pathology reports. TMB-H was defined as > = 10 mutations/megabase. MSI-H was calculated as previously published (PMID: 31570899). Statistical comparisons were limited to groups of 10 or larger.

Results: We characterized a total of 1,477 NENs from the following anatomic sites: lung (64%, N = 951), pancreas (15%, N = 220), GI (15%, N = 224) and prostate (6%, N = 82). TMB-H was most common in NENs arising in the lungs (17%, N = 155), followed by the pancreas (5.4%, N = 11), GI tract (4.8%, N = 10), and prostate (3.8%, N = 3) (p < 0.001). Overall, TMB-H was more common in high-grade NENs compared to low/int-grade NENs [17% (N = 147) vs. 6.2% (N = 31), P < 0.001]. This was also observed in lung NENs [high- grade 19% (N = 135) v low/int grade 10% (N = 19), P = 0.006], with other organ site analyses limited by sample size (although a similar trend was noted for GI NENs). The majority of patients with TMB-H lung NENs had a current or prior smoking history (94%, N = 119). Overall, TMB-H was similar in NENs compared to non-NENs [13% (N = 179) vs. 12% (N = 3018), P = 0.4]. However, in the pancreas, TMB-H was more common in NENs vs. non- NENs [(5.4% (N = 11) v 1.6% (N = 59), P < 0.001] while the reverse was true in lung NENs [17% (N = 155) v 22% (N = 2006), P = 0.001]. MSI-H was observed in only 0.9% (N = 12) of NENs overall, including in 0.4% (N = 4) of lung, 1.4% (N = 3) of pancreas, 1.9% (N = 4) of GI, and 1.2% (N = 1) of prostate NENs. MSI- H status was less common in NENs compared to non-NENs [0.9% (N = 12) vs. 2.5% (N = 614), P < 0.001].

Conclusions: In this real-world NEN cohort, the actionable molecular alterations of TMB-H (13%) and MSI-H (0.9%) were observed. Taken together, the data suggest that molecular profiling (regardless of grade) may help identify a small but meaningful subset of patients with NENs who may benefit from immunotherapeutic treatment approaches. Furthermore, the identification of MSI-H NENs has important implications related to germline testing.

VIEW THE PUBLICATION

VIEW THE POSTER